EP3729241B1 - Areal device offering improved localized deformation - Google Patents

Description

TECHNICAL FIELD AND STATE OF THE PRIOR ART

The present invention relates to a surface device providing improved localized deformation, and in particular to a tactile stimulation interface providing improved stimulation localization.

A tactile stimulation interface is intended to restore tactile information, such as a texture, a relief, a roughness varying in time and/or space, an illusion of pressing a flexible material, of pressing a key, etc.

Such interfaces are used, for example, in the field of human-machine interfaces. They can also be used in the fields of optics, acoustics, chemistry, and automated manufacturing.

A touch interface comprises a surface, for example equipped with a screen. A user interacts with the interface by applying one or more fingers to the interface, for example to make a selection by pressing the representation of a button. On the one hand, we want to be able to realistically reproduce the "click" of selection. On the other hand, we want the user to be able to have several fingers in contact with the interface, to be able to carry out a multi-digital interaction with the interface, and for him to feel sensations that are very distinct from each other.

There are tactile interfaces that implement one or more actuators under the touch surface, for example electromagnetic or piezoelectric actuators, designed to vibrate the entire surface. If several fingers are in contact with the surface, they all perceive the same sensation.

There are also interfaces that include a matrix of actuators, each dedicated to stimulating a specific area of the interface. The finger is either in direct contact with the actuators or isolated by a flexible surface that can deform locally. These devices do not allow for localized feedback on a rigid plate-type surface such as a touchpad (“track pad” in Anglo-Saxon terminology).

By replacing the flexible surface with a rigid plate, for example a glass plate, the vibrations produced locally by an actuator propagate and reverberate throughout the surface. These vibrations are then perceived by all fingers, even if only one actuator under a given finger has been activated.

Furthermore, if several actuators are activated to stimulate several fingers simultaneously, the effects of the vibrations of each actuator are added to those of the other actuator, in the areas where stimulation is actually desired, and also in the other areas. Thus, even the stimulated fingers have "polluted" stimulation.

There are touch interfaces using the time reversal process, for example described in the document C. Hudin, J. Lozada, and V. Hayward, “Localized Tactile Feedback on a Transparent Surface through Time-Reversal Wave Focusing,” IEEE Transactions on Haptics, vol. 8, no. 2, p. 188 198, Apr 2015 . This interface allows stimulation to be localized. The interface comprises a glass plate and actuators arranged in contact with and on the periphery of the glass plate. The piezoelectric actuators propagate acoustic waves in the glass plate. The implementation of a time reversal method has the advantage of allowing a vibration to be generated in a localized manner on the surface of the plate and allows the different fingers to be stimulated separately; this tactile feedback is also called "multitouch localized tactile feedback". This interface is satisfactory, however, this interference requires frequencies typically between 20 kHz and 150 kHz. An audible noise is therefore generated when the finger(s) are in contact with the plate, which produces less natural sensations than low-frequency vibrations such as those produced by the mechanical response of a keyboard key. Indeed, low-frequency vibrations, in the touch sensitivity range, generally between 0 kHz and 1 kHz, allow, when correlated with a user action, such as force, movement, contact, to simulate the presence of keys or relief on the surface of a screen. A similar device is disclosed by US2011/090167A1 .

The document H. Nicolau, K. Montague, T. Guerreiro, A. Rodrigues, and VL Hanson, "HoliBraille: multipoint vibrotactile feedback on mobile devices," 2015, pp. 1-4 proposes to solve the problem by placing an actuator under each finger and mechanically isolating each actuator with a vibration-absorbing surface. Activating an actuator then produces a stimulus perceived only by the finger in direct contact. In this case, the finger is therefore in direct contact with the actuator and not with the touch surface with which it interacts. Furthermore, sufficiently damping low-frequency vibrations requires a large volume of foam, which is not compatible with the space constraints in a mobile device, such as a touchscreen tablet or a smartphone.

STATEMENT OF THE INVENTION

The invention is defined in the independent claims. The dependent claims define advantageous embodiments. It is therefore an object of the present invention to provide a surface device offering improved localization of the deformation of its surface.

It is also an aim of the present invention to provide a tactile stimulation interface allowing the generation of improved stimulation localization, while presenting a smooth surface.

The aim stated above is achieved by a tactile stimulation interface comprising a surface intended to be explored tactilely by a user and at least one actuator intended to generate a vibration at an area of the surface where it is desired to generate a tactile stimulation, means for controlling said at least one actuator, comprising means for calculating the control signals implementing an inverse filtering operation, and sending control signals to the actuator. The inverse filtering makes it possible to compensate for the propagation effects.

In an embodiment implementing several actuators each located under an area of the surface, when it is desired to stimulate a finger located above one of the actuators, the actuator is activated, but the other actuators located under the fingers that are not desired to be stimulated are also activated so as to cancel the vibrations in these areas. In addition, the actuator activated to stimulate actually a finger is controlled taking into account the effect of the activation of the other actuators.

In another embodiment of an interface implementing several actuators, the finger(s) are not located above one or more actuators and when it is desired to stimulate a finger, all the actuators receive a signal to, on the one hand, generate a vibration of the area of the surface in contact with the finger to be stimulated and, on the other hand, cancel the vibrations in the areas of the surface in contact with the other fingers that are not desired to be stimulated. In this embodiment, the control points are not co-located with the actuators.

In other words, the invention does not prevent the transmission or propagation of waves across the entire surface but cancels, thanks to the various actuators, the vibrations at the points where stimulation is not desired. The actuators are therefore used both to produce vibrations intended to generate a desired stimulation and to cancel vibrations.

The invention therefore makes it possible to compensate for the reverberation of the waves and their propagation which cause a cross-talk phenomenon, i.e. pollution of the desired movement at a given point by the signal sent to another actuator at another point on the plate.

The use of this inverse filter therefore makes it possible to obtain, in different areas of the surface, whether actuators are located under these areas, part of them or under none of them, a displacement corrected for the effects of dispersion and reverberations.

Thanks to the invention, the movement obtained in the area where the finger to be stimulated is located is independent of the movements obtained at the center of the other actuators. The other fingers are then not stimulated. It is then possible to create a multi-digital tactile interface, thanks to a partitioning of the stimulations from one area to another.

It is also possible to consider the case of an interface to an actuator. Indeed, the generation of the control signals of the single actuator compensates for signal distortion due to actuator response and wave reverberation in the surface.

The touch interface has a footprint suitable for touch applications, its volume is not increased compared to existing interfaces. Indeed, the actuators can be glued directly to the touch surface and vibration damping is no longer necessary, the implementation of bulky isolation means is therefore no longer required.

The present invention thus relates to a surface device with localized deformation as defined by claim 1.

Preferably, the interaction surface comprises several interaction zones arranged relative to each other, so that they cover substantially the entire interaction surface and at least as many actuators as interaction zones, said calculation means implementing an inverse filtering operation, so as to emit, from one or more desired movements of one or more interaction zones, control signals at least partially compensating for the distortion, reverberation and propagation of the waves.

In an exemplary embodiment, the actuator(s) are arranged under said interaction zone(s), opposite the interaction surface.

Preferably, the surface of the actuator(s) corresponds substantially to that of the interaction element(s) intended to come into contact with the interaction surface.

In the case where the elements are fingers, the surface area of the actuator(s) is advantageously between 1 cm ² and a few cm ² .

In another exemplary embodiment, the interaction zone(s) are distant from the actuator(s) in the plane of the interaction surface.

Very advantageously, the device comprises means for detecting contact between at least the interaction zone and an external interaction element, and preferably, means for detecting contact between the external interaction element(s) and all the interaction zones.

The device may include means for measuring the pressure force of the external element(s) with the interaction zone(s) to determine the desired stimulation.

For example, the interaction zones and actuators have a hexagon shape, which helps to optimize the coverage of the interaction surface.

In one exemplary embodiment, the actuators are piezoelectric actuators. The actuators may comprise transparent thin films making them suitable for the manufacture of touch screens.

In another exemplary embodiment, the actuators are electromagnetic actuators each comprising a coil and a magnet, the magnet or the coil being capable of exerting a force on the plate.

In one exemplary embodiment, at least a portion of each actuator is attached directly to the plate.

The device may include a screen disposed under the plate opposite the interaction surface. The screen may be attached to the plate opposite the interaction surface. The actuators may be attached to the screen opposite the face of the screen in contact with the plate.

The present invention also relates to a tactile stimulation interface or a touchpad as defined by claim 12.

• détection d'un contact entre ladite zone d'interaction et l'élément d'interaction extérieur,
• choix d'un déplacement souhaité de ladite zone d'interaction,
• génération d'un signal de contrôle par une opération de filtrage inverse à partir du déplacement souhaité,
• application du signal de contrôle audit actionneur.

The present disclosure further relates to a method for operating a surface device with localized deformation comprising a plate carrying an interaction surface with one or more external interaction elements, comprising at least one zone of interaction with the exterior, at least one actuator in contact with the interaction surface and capable of causing deformation in a direction orthogonal to the plane of the plate, comprising:

• detection of contact between said interaction zone and the external interaction element,
• choice of a desired displacement of said interaction zone,
• generation of a control signal by an inverse filtering operation from the desired displacement,
• applying the control signal to said actuator.

The present invention also relates to the method of operating a surface device with localized deformation defined by claim 13.

In one example of operation, some or all of the actuators are arranged under the interaction zones, and control signals are applied to some or all of the actuators located under an interaction zone with which contact with an external element has been detected.

BRIEF DESCRIPTION OF THE DRAWINGS

• la figure 1 est une vue de dessus d'une représentation schématique d'une interface tactile mettant en œuvre quatre actionneurs, selon un premier mode de réalisation de la présente invention,
• la figure 2 est une représentation du déplacement souhaité en µm en fonction du temps en ms,
• la figure 3A représente graphiquement les signaux émis en V en fonction du temps en ms, pour les actionneurs A1 à A4 pour une interface de l'état de la technique en vue d'obtenir le déplacement de la figure 2,
• la figure 3B représente graphiquement les signaux émis en V en fonction du temps en ms, pour les actionneurs A1 à A4 pour une interface selon l'invention en vue d'obtenir le déplacement de la figure 2,
• les figures 4A sont des représentations graphiques du déplacement mesuré au centre d'une zone au-dessus de chaque actionneur, en actionnant les actionneurs A1 à A4 de l'interface de la figure 1 avec les signaux de la figure 3B,
• les figures 4B à 4D sont des représentations graphiques du déplacement mesuré au centre des zones Z1 à Z3 lorsque le déplacement de la figure 2 est souhaité dans les zones Z2, Z3, Z4 respectivement, en actionnant les actionneurs A2, A3, A4 de l'interface de la figure 1,
• les figures 5A sont des représentations graphiques du déplacement mesuré au centre d'une zone au-dessus de chaque actionneur, en actionnant les actionneurs A1, A2, A3 et A4 avec les signaux de la figure 3A,
• les figures 5B à 5D sont des représentations graphiques du déplacement mesuré au centre des zones au-dessus de chaque actionneur, lorsque le déplacement souhaité de la figure 2 est souhaité dans les zones Z2, Z3, Z4 respectivement, en actionnant les actionneurs A2, A3, A4 de l'interface de l'état de la technique,
• la figure 6 est une représentation schématique des étapes de fonctionnement d'une interface tactile selon un autre exemple de la présente invention,
• les figures 7A à 7D sont des vues en coupe selon un plan orthogonal au plan de la surface tactile de différents exemples de réalisation de structures d'une interface tactile selon la présente invention,
• la figure 8 est une vue de dessus d'une interface tactile selon un deuxième mode de réalisation selon l'invention,
• la figure 9A est une représentation schématique de deux interfaces selon l'invention avec des dispositions matricielles des actionneurs de deux tailles différentes,
• la figure 9B est une représentation graphique du ratio d'énergie en dB de l'ensemble des signaux de pilotage nécessaire au contrôle de deux points disposés aléatoirement sur la plaque pour les deux interfaces de la figure 9A,
• la figure 10A est une représentation schématique de deux interfaces de l'état de la technique dans lesquelles les actionneurs de deux tailles différentes sont disposés sur les bords de l'interface,
• la figure 10B est une représentation graphique du ratio d'énergie en dB de l'ensemble des signaux de pilotage nécessaire au contrôle de deux points disposés aléatoirement sur la plaque pour les deux interfaces de la figure 10A.

The present invention will be better understood on the basis of the following description and the attached drawings in which:

• there Figure 1 is a top view of a schematic representation of a touch interface implementing four actuators, according to a first embodiment of the present invention,
• there Figure 2 is a representation of the desired displacement in µm as a function of time in ms,
• there Figure 3A graphically represents the signals emitted in V as a function of time in ms, for actuators A1 to A4 for a state-of-the-art interface in order to obtain the displacement of the Figure 2 ,
• there Figure 3B graphically represents the signals emitted in V as a function of time in ms, for the actuators A1 to A4 for an interface according to the invention in order to obtain the displacement of the Figure 2 ,
• THE Figures 4A are graphical representations of the displacement measured at the center of an area above each actuator, by actuating actuators A1 to A4 of the interface of the Figure 1 with the signals of the Figure 3B ,
• THE Figures 4B to 4D are graphical representations of the displacement measured at the center of zones Z1 to Z3 when the displacement of the Figure 2 is desired in zones Z2, Z3, Z4 respectively, by actuating actuators A2, A3, A4 of the interface of the Figure 1 ,
• THE Figures 5A are graphical representations of the displacement measured at the center of an area above each actuator, by operating actuators A1, A2, A3 and A4 with the signals from the Figure 3A ,
• THE Figures 5B to 5D are graphical representations of the displacement measured at the center of the zones above each actuator, when the desired displacement of the Figure 2 is desired in zones Z2, Z3, Z4 respectively, by actuating actuators A2, A3, A4 of the state-of-the-art interface,
• there Figure 6 is a schematic representation of the steps of operation of a touch interface according to another example of the present invention,
• THE Figures 7A to 7D are sectional views along a plane orthogonal to the plane of the touch surface of different exemplary embodiments of structures of a touch interface according to the present invention,
• there figure 8 is a top view of a touch interface according to a second embodiment according to the invention,
• there Figure 9A is a schematic representation of two interfaces according to the invention with matrix arrangements of the actuators of two different sizes,
• there Figure 9B is a graphical representation of the energy ratio in dB of all the driving signals required to control two points randomly arranged on the plate for the two interfaces of the Figure 9A ,
• there Figure 10A is a schematic representation of two state-of-the-art interfaces in which actuators of two different sizes are arranged on the edges of the interface,
• there Figure 10B is a graphical representation of the energy ratio in dB of all the driving signals required to control two points randomly arranged on the plate for the two interfaces of the Figure 10A .

DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

In the following description, the invention will be described more particularly in an application to a touch interface, but the present invention applies to other fields such as, for example, micromanipulation or optics.

In the case of a touch interface, it will be considered that the user interacts with the touch interface with his fingers. It will be understood that he could interact with other parts of his body.

In this application, the expressions "interaction zone", "stimulation zone" and "control point" are synonymous.

On the Figure 1 , we can see a schematic representation of a top view of a first embodiment of a touch interface according to the invention, comprising a plate 1, for example made of glass, carrying on one of its faces the surface for interaction with the exterior, designated touch surface, four actuators A1, A2, A3 and A4 arranged under the glass plate, for example fixed on the surface of the plate 1 opposite the touch surface 2. The touch interface also comprises means 6 for controlling each of the actuators comprising means 8 for calculating control signals.

The plate material is chosen such that it allows low frequency vibrations, typically less than <1 kHz, to propagate over a few cm. The material can be a flexible or rigid material.

The actuators are such that they are capable, when activated, of exerting a force on the plate in an out-of-plane direction, i.e. orthogonal to the plane of the plate. The plane of the plate is the plane extending parallel to its largest surface. On the Figures 7A to 7D , several examples of interface according to the invention can be seen from the side. The actuators are capable of exerting an upward and/or downward force in the representation of the Figures 7A to 7D .

As will be described later, the actuators may or may not be in direct contact with the plate.

In the example shown, the actuators are aligned along an axis.

Actuators are for example piezoelectric actuators.

The user is intended to interact with the touch-sensitive surface 2, for example by pressing on certain locations on the surface designated Z1, Z2, Z3, Z4. The actuators A1 to A4 are arranged directly above the zones Z1 to Z4 respectively and are intended to be actuated to tactilely stimulate the fingers in contact with the zones. The surface area of the zones Z1 to Z4 is equal to the surface area of the actuators A1 to A4.

Preferably, the surface area of the actuators corresponds to the surface area with which the fingers come into contact with the touch surface, so that only one finger at a time is in contact with an area Z1 to Z4. For example, the external dimensions of an actuator are of the order of cm, for example a disc with a diameter of the order of cm or a square with a side of the order of cm. Thus the surface area of the actuators by which they will act on the plate is preferably between 1 cm ² and a few cm ² .

For example, the actuators have a disc shape with a diameter of Φ 2 cm, and for example, the touch surface has a length of 15 cm and a width of 10.5 cm.

As will be described later, the invention can activate the actuators so as to control the tactile stimulation of each area located above an actuator. The larger the surface area covered by the actuators, the better the control of the tactile stimulation on the tactile surface. The actuators can have any shape, such as a disc or polygon...For example, the actuators are hexagonal in shape so as to ensure maximum paving under the tactile surface as shown in the Figure 6 .

The calculation means 8 implement an inverse filtering operation to determine the control signals. The calculation means also implement a vibration synthesis algorithm determining the desired signal in a zone, as a function of the desired stimulation in this zone, and taking into account for example the pressure force on this zone, the speed of movement of the fingers as will be described below. This type of algorithm is well known to those skilled in the art and will not be described in detail.

We will describe an example of operation of a touch interface without calculation means 8 applying an inverse filtering operation, and with the calculation means 8 applying the inverse filtering operation. In this example we wish to obtain the movements in the zones Z1 to Z4 represented on the Figure 2 , which were determined by the vibration synthesis algorithm. In zone Z1, we want a windowed sinus type displacement "burst" of 5 oscillation cycles at 300 Hz, and no displacement in zones Z2 to Z4, i.e. no vibration.

On the Figure 3A , we can see the control signals in Volt as a function of time in ms emitted and sent to actuators A1, A2, A3 and A4 in a state-of-the-art touch interface. Only a signal is sent to actuator A1 which is identical to the desired displacement and no signal is sent to actuators A2, A3 and A4.

On the Figures 5A , we can see the displacements measured in µm as a function of time in ms in zones Z1, Z2, Z3 and Z4 resulting from the signals of the Figure 3A .

It can be seen that, on the one hand, the displacement measured in zone Z1 corresponds to the deformed control signal and presents additional oscillations due to the reflections of the waves produced by the plate and their propagation, it therefore does not correspond to the desired displacement represented on the Figure 2 .

On the other hand, we note that non-zero displacements are measured in zones Z2 to Z4, while no displacement was desired in these zones. Furthermore, these displacements are not negligible. Thus, if a user has one or more fingers on zones Z2, Z3 and/or Z4, he will feel an unwanted tactile stimulation. He may then perceive false information.

Graphical representations of the Figures 5B, 5C and 5D show the displacements measured in all areas when activating actuators A2, A3 and A4 respectively by applying the signal from the Figure 3A which was applied to A1.

We therefore note that by applying a control signal which corresponds directly to the desired movement, there is on the one hand a difference between the desired movement in an area and the movement obtained, and on the other hand that unwanted tactile stimulations are generated.

According to the invention, the calculation means implement inverse filtering, which makes it possible to better control the movements in each of the zones, whether these movements are zero or not.

On the Figure 3B , we can see the control signals emitted and sent to the actuators A1, A2, A3 and A4 in a touch interface according to the invention to obtain the desired movement of the Figure 2 . All actuators are activated and not just actuator A1 and the signal sent to A1 is not identical to the desired displacement, it is complex and is such that it compensates for the effects of the other actuators and the reflection effects.

On the Figure 4A , we can see the measured displacements resulting from the signals of the Figure 3B in each zone Z1 to Z4 We note that the measured displacement in Z1 corresponds to the desired one and that the movements measured in the other zones where no movement is desired are almost zero. Thus, if the user places his finger on one of the zones Z2 to Z4, he does not feel any tactile stimulation or only very little. The information transmitted to the user is therefore correct.

THE Figures 4B to 4D show the displacements measured in all areas when activating actuators A2, A3 and A4 respectively by applying the signal from the Figure 3B which was applied to A1. We note that the displacements measured in Z2, Z3 and Z4 correspond to those desired, and that the displacements measured in the other zones where no displacement is desired are almost zero.

By means of the inverse filtering operation, the control signals are such that for the areas for which no stimulation is desired, they activate the actuators corresponding to these areas, at least those under the areas with which a finger is in contact, so that it generates vibrations aimed at cancelling those resulting by propagation of the activation of the actuator under the area where stimulation is desired to be generated.

The calculation of the actuator control signal under the area where stimulation is to be generated takes into account both the desired displacement and the effect of the propagation and reflection of the vibrations produced by the other actuators. According to the invention, each actuator is therefore controlled taking into account the external environment.

We can then obtain, in each zone covered by an actuator, a displacement which can be zero, corrected for distortion and reverberation effects, and independent of the displacements in the center of the other zones.

We will describe the inverse filtering operation. Such an operation is for example described in the article “Optimal focusing by spatio-temporal inverse filter. I. Basic principles » M.Tanter et al., The Journal of the Acoustical Society of America 110, 37 (2001 ) applied to image processing in medical imaging.

The response R of a linear system to an excitation E is given by the relation R=HE, with H the transfer function of the system. In the application to a touch interface, we observe the displacement U _i of the plate measured at the center of an actuator i in response to a signal S _j sent to an actuator j. We therefore have: $${\mathit{U}}_{\mathit{i}}\left(\omega \right)={\mathit{H}}_{\mathit{ij}}\left(\omega \right){\mathit{S}}_{\mathit{j}}\left(\omega \right)$$

With H _ij (ω) the transfer function between the signal sent to actuator i and the displacement recorded at the center of actuator j. If N actuators emit simultaneously, the displacement obtained is the sum of the contributions of these N actuators, i.e.: $${\mathit{U}}_{\mathit{j}}\left(\omega \right)={\displaystyle \sum _{\mathit{i}=1}^{\mathit{N}}{\mathit{H}}_{\mathit{ij}}\left(\omega \right){\mathit{S}}_{\mathit{j}}\left(\omega \right)}$$

ou $${\mathbb{U}}_{\omega }={ℍ}_{\omega }.{\mathbb{S}}_{\omega }$$

In matrix form we write: $${\left[\begin{array}{c}{\mathit{U}}_1\\ {\mathit{<figure-callout\; id="2"\; label="U"\; filenames="imgf0003.png,imgf0004.png"\; state="\{\{state\}\}">U</figure-callout>}}_2\\ ⋮\\ {\mathit{U}}_{\mathit{N}}\end{array}\right]}_{\omega }={\left[\begin{array}{cccc}{\mathit{H}}_{11}& {\mathit{<figure-callout\; id="12"\; label="H"\; filenames="imgf0005.png,imgf0008.png"\; state="\{\{state\}\}">H</figure-callout>}}_{12}& \cdots & {\mathit{H}}_{1\mathit{N}}\\ {\mathit{H}}_{21}& {\mathit{<figure-callout\; id="22"\; label="H"\; filenames="imgf0001.png,imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}_{22}& \cdots & {\mathit{H}}_{2\mathit{N}}\\ ⋮& ⋮& \ddots & ⋮\\ {\mathit{H}}_{\mathit{N}1}& {\mathit{<figure-callout\; id="2"\; label="H\; N"\; filenames="imgf0003.png,imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathit{N}}& \cdots & {\mathit{H}}_{\mathit{NN}}\end{array}\right]}_{\omega }\cdot {\left[\begin{array}{c}{\mathit{S}}_1\\ {\mathit{<figure-callout\; id="2"\; label="S"\; filenames="imgf0003.png,imgf0004.png"\; state="\{\{state\}\}">S</figure-callout>}}_2\\ ⋮\\ {\mathit{S}}_{\mathit{N}}\end{array}\right]}_{\omega }$$

Or $${\mathbb{U}}_{\omega }={ℍ}_{\omega }.{\mathbb{S}}_{\omega }$$

The displacement u _i at the center of an actuator i is therefore not proportional to the signal s _i applied to it but is filtered by the response of the actuator stuck to the plate H _ii and depends, via the terms H _ij , on the signals sent to the other actuators which produce waves propagating throughout the plate.

le déplacement souhaité, dans le domaine fréquentiel, à l'ensemble des positions, on calcule le signal ${\mathbb{S}}_{\omega }={\left[\begin{array}{c}{\mathit{S}}_1\\ {\mathit{S}}_2\\ ⋮\\ {\mathit{S}}_{\mathit{N}}\end{array}\right]}_{\omega }$

à envoyer à chacun des actionneurs par la relation : $${\mathbb{S}}_{\omega }={ℍ}_{\omega }^{-1}.{\mathbb{V}}_{\omega }$$

Inverse filtering involves inverting this relationship by calculating the signal to be applied to all the actuators to obtain the desired displacement. Noting ${\mathbb{V}}_{\omega }={\left[\begin{array}{c}{\mathit{V}}_1\\ {\mathit{<figure-callout\; id="2"\; label="V"\; filenames="imgf0003.png,imgf0004.png"\; state="\{\{state\}\}">V</figure-callout>}}_2\\ ⋮\\ {\mathit{V}}_{\mathit{N}}\end{array}\right]}_{\omega }$

the desired displacement, in the frequency domain, at all positions, we calculate the signal ${\mathbb{S}}_{\omega }={\left[\begin{array}{c}{\mathit{S}}_1\\ {\mathit{<figure-callout\; id="2"\; label="S"\; filenames="imgf0003.png,imgf0004.png"\; state="\{\{state\}\}">S</figure-callout>}}_2\\ ⋮\\ {\mathit{S}}_{\mathit{N}}\end{array}\right]}_{\omega }$

to send to each of the actuators by the relation: $${\mathbb{S}}_{\omega }={ℍ}_{\omega }^{-1}.{\mathbb{V}}_{\omega }$$

donné par : $${\mathbb{U}}_{\omega }={ℍ}_{\omega }.{\mathbb{S}}_{\omega }={ℍ}_{\omega }.{ℍ}_{\omega }^{-1}.{\mathbb{V}}_{\omega }={\mathbb{V}}_{\omega }$$

We finally get a displacement ${\mathbb{U}}_{\omega }$

given by: $${\mathbb{U}}_{\omega }={ℍ}_{\omega }.{\mathbb{S}}_{\omega }={ℍ}_{\omega }.{ℍ}_{\omega }^{-1}.{\mathbb{V}}_{\omega }={\mathbb{V}}_{\omega }$$

.This gives a movement that conforms to that expected. ${\mathbb{U}}_{\omega }={\mathbb{V}}_{\omega }$

.

By inverting the matrix, all the effects are compensated, before generating the control signals to obtain the desired displacement despite distortions, reverberations and wave propagations.

This filter is temporal insofar as it operates a transformation on the amplitude and phase at all frequencies, and spatial since it takes into account the signals emitted by all the actuators.

Preferably, the interface comprises detection means 10 for detecting the presence of a finger on an area in order, on the one hand, to determine whether stimulation is to be generated and, on the other hand, to activate the actuators under the areas which must not be activated. The detection means implemented are those usually implemented in the tactile domain, for example they are of the capacitive, resistive, infrared type, etc. Alternatively, it is possible to simply detect the finger on the stimulation area and control the actuators of all the other areas in order to limit or even cancel their movement. However, this activation consumes energy and computing power.

According to another variant, the interface does not include means for detecting the presence of a finger, so it is possible to produce a vibration in an area without knowing whether a finger is actually on this area. Vibration control is then carried out by assuming that all positions are touched.

Advantageously, the interface comprises means 11 for measuring the pressure force of the fingers on the zones, the value of the pressure force can then advantageously be taken into account to more faithfully simulate the response of a key or a button. The means for measuring the force are for example piezoelectric, piezoresistive, capacitive, etc. the pressure force value is taken into account by the vibration synthesis algorithm to determine the desired displacement in a zone, and not during the inverse filtering step.

Also advantageously, the speed of the finger(s) on the touch surface is also measured and taken into account by the vibration synthesis algorithm to determine the shape of the signal that is desired.

By taking into account the force of support and the speed of the movements, the stimulation is then more realistic.

On the Figure 6 , we can see a schematic representation of an example of the operation of the calculation means and the control means of another example of a touch interface.

In this example, the interface comprises a plurality of hexagonal actuators A1, A2, A3...AN covering almost the entire face opposite the touch surface. Thus, regardless of the position of the fingers on the touch surface, the movement of the area with which a finger is in contact can be controlled by activating the actuator located under this area.

On the Figure 6 , three tactile stimulations are produced in three distinct zones, these can be identical or different. Three fingers D1, D2, D3 are in contact with the tactile surface in three distinct zones Z1, Z2, Z3. Indeed the calculation of the control signals can be done when one wishes to stimulate in at least two distinct zones. In this operating mode, the displacement of zones is kept zero except zones Z1, Z2 and Z3. The calculation of the control signals is carried out by applying an inverse filtering.

The presence of fingers on zones Z1, Z2 and Z3 is detected and possibly their pressure force on zones Z1, Z2 and Z3 is measured.

Each zone is or are associated with one or more stimulations stored in a memory of the control means, this stimulation can vary depending for example on the pressing force. For example, the stimulation can be such that it reproduces the movement of a keyboard key being pressed, of a validation button, click type; the transient vibrations produced when pressing on a deformable surface can also be reproduced. It can be seen that the screen has patterns corresponding to different commands.

The control means synthesize the information collected (step 200), and then determine the desired vibration (step 300) which was associated with a stimulation during the programming of the interface.

The desired vibrations then serve as input (step 400) to the inverse filter of the calculation means 8 which determine the control signals at least of the actuators A1, A2 and A3 (step 500). The signals are advantageously amplified and are then sent to the actuators A1, A2 and A3 (step 600). They then produce a compliant vibrotactile feedback (step 700).

The interface according to the invention may comprise only one actuator; in fact, the calculation of the control signal of the single actuator by inverse filtration makes it possible to compensate for the distortion of the signal due to the response of the actuator to its own vibration and to the reverberation of the waves in the surface.

The interface according to the invention allows working at all frequencies and not only at touch sensitivity frequencies below 1 kHz, however these are advantageous because they do not produce sound when activating the actuators. Thus different types of actuators can be used. Piezoelectric actuators are suitable for high and low frequency operation.

A piezoelectric actuator comprises a piezoelectric material in the form of a plate, for example PZT (Lead Zirconate Titanium) or AlN (Aluminum Nitride), and electrodes on either side of the plate and in contact with it, to apply a current to it to cause the deformation of the piezoelectric material.

Thanks to the invention, it is also possible to give a controlled profile to the surface. Indeed, a permanent deformation of the surface can be seen as a vibration of zero frequency. The inverse filter method can therefore be applied. By exerting a localized force on a plate, the entire surface is deformed. By applying the inverse filter method, this deformation can be canceled at the desired points.

Electromagnetic actuators are possible. They are suitable for low-frequency operation. Such actuators are, for example: described in the document Benali-Khoudja et al. - 2007 - VITAL An electromagnetic integrated touch display ". For example, the actuators each have a fixed coil and a magnet glued under the touch surface. The current signal sent to the coils is calculated by inverse filtering.

On the Figures 7A to 7D , several examples of touch interface structures applicable to the present invention can be seen.

There Figure 7A , the actuators A1, A2...AN are fixed, for example by gluing directly onto the face of the plate 1 opposite the touch surface 2. This structure is suitable for the production of a touch pad, since the actuators are generally not transparent. In the case of a touch pad, the touch surface is generally opaque.

On the Figure 7B , the interface comprises a plate 1 provided with actuators as in Figure 6A, and a screen E opposite the touch plate. In this structure, transparent actuators are advantageously chosen, for example piezoelectric actuators deposited in a thin layer.

On the Figure 7C , the interface comprises a transparent plate 1, a screen E arranged directly under the plate 1 and secured to it, it is for example glued to the plate 1, and actuators A1...AN fixed on the screen on the face opposite to that oriented towards the side of the plate. This configuration has the advantage of not requiring transparent actuators. In this example, the actuators act on the touch surface through the screen. The screen is for example an OLED screen which has the advantage of being very thin and is generally glued directly to the touch plate. This assembly has the advantage of offering good transmission of low frequencies

On the Figure 7D , the interface comprises a plate 1 and piezoelectric actuators A1 to AN on the face opposite the touch surface. The actuators comprise in common a layer of piezoelectric material 12, a common electrode 14 between the layer 12 and the plate 1 and electrodes 16 on the face opposite the layer 12 so as to produce individual actuators.

On the figure 8 , we can see a top view of a touch interface according to a second embodiment.

In this embodiment, the finger(s) to be stimulated, and therefore the areas of the surface to be stimulated, are not located above actuators.

In this example, actuators A101 to A106 are distributed along the edges of the touch surface, three on each edge. The fingers are intended to come into contact with areas of the surface located between the two rows of actuators. These areas Z101, Z102, Z103... are potential stimulation areas. The arrangement of the actuators of the figure 8 is not limiting, any other arrangement is possible, for example a distribution along the four edges of the plate, or on two non-parallel edges, a non-symmetrical distribution, a distribution in a circle in the case of a circular surface...

The actuators can be arranged on or under the surface, for example glued to the surface.

This embodiment is very advantageous in a single-screen application because it does not require the implementation of transparent actuators.

Preferably, the actuators are arranged under the entire interaction surface. This arrangement makes it possible to minimize the distance between the control points where the fingers can be located, and the actuators.

The potential stimulation zones are located in the near field of the actuators, i.e. the potential stimulation zones are located at a distance less than or equal to the dimension of the actuators in the plane or to the wavelength of the control signals sent to the actuators, whichever is greater.

This near-field configuration allows efficient control by reducing the power of the emitted signals to obtain given displacements, especially when the control points are less than one wavelength apart.

On the Figure 9B , we can see the energy ratio in dB of all the control signals necessary to control two points randomly arranged on the plate for the matrix arrangements of the actuators A of two different sizes schematized on the Figure 9A allowing the control points to be placed in the near field of the actuators according to the invention

For comparison, on the Figure 10B , we can see the energy ratio in dB of all the control signals necessary for controlling two points randomly arranged on the plate for the arrangements of the actuators A' of two different sizes located on the edge of the plate shown diagrammatically on the Figure 10A , and for which the control points are not in the near field of the actuators

These measurements were made using a 1 mm thick glass plate with a frequency band of the control signals covering the 0-1 kHz touch sensitivity range which corresponds to wavelengths of 10 cm. The actuators considered are piezoelectric ceramics of dimensions 10 mm and 20 mm. 1000 pairs of points chosen randomly but identical for the two simulations were tested. An average reduction of 5 and 8 dB respectively is observed for the two sizes of actuators used thanks to the relative arrangement of the actuators and the control points according to the invention, compared to an arrangement of the actuators on the edges of the plate.

The interface also comprises control means 106 comprising calculation means 108 implementing an inverse filtering operation, in which the matrix grouping the transfer functions between the signal sent to each actuator and the movements recorded in the different potential stimulation zones, may not be a square matrix since the number of actuators and the number of potential stimulation zones may be different. In order to ensure the stability of the matrix inversion, the number of actuators is greater than or equal to the maximum number of zones to be stimulated simultaneously, i.e. in the case of an interface used with one or two hands the number of actuators is greater than or equal to the maximum number of fingers that can come into contact with the surface, 5 or 10 for example.

Preferably, the interface comprises means for detecting finger contact on the different areas of the surface.

As for the first embodiment, the control of the actuators uses a matrix H ( ω ) established from the frequency response functions H _pq (ω) linking the Q actuators to each of the P fingers.

These frequency response functions can be obtained from a database of responses or interpolated from a reduced database of responses.

qui est une pseudo-inverse de la matrice ${ℍ}_{\omega }$

, car la matrice peut ne pas être carrée, pour chaque fréquence de la bande passante.We then calculate the matrix ${ℍ}_{\omega }^{-1}$

which is a pseudo-inverse of the matrix ${ℍ}_{\omega }$

, because the matrix may not be square, for each frequency in the bandwidth.

The operating mode of the control means is as follows, considering an interface with Q actuators and with P fingers likely to come into contact with the surface of the interface.

First of all, the position of the finger(s) on the interaction surface is determined by detection means similar to those described above in relation to the first embodiment.

Depending on the type of interaction, all or part of the fingers on the surface are stimulated. In a next step, the desired vibrations v _p ( t ) are determined under each of the P fingers. These vibrations are arbitrary signals previously determined according to the information to be provided, possibly zero, determined so as to produce haptic feedback perceptible by the user and adapted to the interaction context.

pour obtenir les signaux de pilotage des actionneurs.In a next step, the desired vibrations are filtered using the inverse matrix ${ℍ}_{\omega }^{-1}$

to obtain the actuator control signals.

In a next step, the control signals s _q ( t ) of Q actuators are emitted and sent to the actuators.

For example, we want finger D1 to be stimulated and the other fingers D2 and D3 not to be stimulated. All actuators A101 to A06 are driven to generate stimulation in zone Z101 and to counter any vibration that may appear in zones Z102 and Z103 and to optimize stimulation in zone Z101.

The operating mode of the interface according to the second embodiment is close to that of the interface according to the first mode.

As for the first embodiment, the stimulations to be generated can be modulated for example according to the pressure force of the finger(s) on the surface. Alternatively, the device may not include means for detecting the contact(s) of one or more fingers or other limbs.

Alternatively, the interface has a single actuator.

In another embodiment, the interface is such that the potential stimulation zones are located above the actuators or not. The number of actuators is chosen to be greater than the number of potential stimulation zones.

The present invention is particularly suitable for human-machine interaction with a touch surface. The present invention can also be implemented in applications in adaptive optics or micromanipulation, which requires high control of the deformations and vibrations of a surface.

The present invention also applies to interfaces whose surface is not flat, i.e. it applies to interfaces comprising complex curved surfaces, for example of the shell type.

Claims (13)

1. Surface device with localized deformation, including a plate carrying a surface for interaction with one or more external interaction elements, including zones for interaction with the exterior, actuators able to cause a deformation in a direction orthogonal to the plane of the plate at the interaction zones of the means for detecting contact between the interaction zones and an external interaction element, means for controlling the actuators configured to send control signals to the actuators, comprising means for computing said control signals, said computing means implementing an inverse filtering operation, so as to emit, from a required movement of at least one of the interaction zones, control signals at least partially compensating for the distortion, the reverberation and the propagation of the waves, the device being characterized in that each of the interaction zones is in a near field of at least one of the actuators while being located at a distance less than or equal to the wavelength of the control signals sent to the actuators and/or less than or equal to the dimensions of the actuators in the directions of the interaction surface.
2. Device according to claim 1, wherein the interaction zones are disposed with respect to one another so that they cover substantially the entire interaction surface and the actuators, said computing means implementing an inverse filtering operation, so as to emit, from one or more required movements of one or more interaction zones, control signals at least partially compensating for the distortion, the reverberation and the propagation of the waves, for example the interaction surface including a plurality of interaction zones disposed with respect to one another so that they cover substantially the entire interaction surface and at least as many actuators as there are interaction zones, said computing means implementing an inverse filtering operation, so as to emit, from one or more required movements of one or more interaction zones, control signals at least partially compensating for the distortion, the reverberation and the propagation of the waves.
3. Device according to claim 1 or 2, wherein the surface area of the actuators is between 1 cm² and a few cm².
4. Device according to one of claims 1 to 3, wherein the actuators are disposed under said interaction zones, opposite to the interaction surface.
5. Device according to one of claims 1 to 4, wherein the interaction zones are distant from the actuator or actuators in the plane of the interaction surface.
6. Device according to one of claims 1 to 5, including means for detecting the contact between the external interaction elements and all the interaction zones.
7. Device according to one of claims 1 to 6, including means for measuring the bearing force of the external element or elements with the interaction zones.
8. Device according to one of claims 1 to 7 taken in combination with claim 2, wherein the interaction zones and the actuators have a hexagon shape.
9. Device according to one of claims 1 to 8, wherein the actuators are piezoelectric actuators, the actuators advantageously including transparent thin films.
10. Device according to one of claims 1 to 8, wherein the actuators are electromagnetic actuators each including a coil and a magnet, the magnet or the coil being able to exert a force on the plate.
11. Device according to one of claims 1 to 10, including a screen disposed under the plate opposite to the interaction surface, the screen advantageously being secured to the plate opposite to the interaction surface, the actuators advantageously being secured to the screen opposite to the face of the screen in contact with the plate.
12. Touch stimulation interface including a device according to one of claims 1 to 11 or touch pad including a device according to one of claims 1 to 10, wherein at least part of each actuator is secured directly to the plate.
13. Method for operating a surface device with localized deformation including a plate carrying a surface for interaction with one or more external elements, including a plurality of zones for interaction with the exterior, a plurality of actuators in contact with the interaction surface and able to cause a deformation in a direction orthogonal to the plane of the plate,

    the method including:

    - detecting one or more contacts between said interaction zones and the interaction elements,

    - selecting a required movement for each of said interaction zones,

    - generating control signals by an inverse filtering operation from the required movements,

    - applying control signals to at least one of the actuators,

    and advantageously wherein all or some of the actuators are disposed under the interaction zones and wherein control signals are applied to all or some of the actuators located under an interaction zone with which a contact with an external element has been detected,

    the method being characterized in that each of said interaction zones is located in a near field of at least one of the actuators while being located at a distance less than or equal to the wavelength of control signals sent to said at least one actuator and/or less than or equal to the dimensions of said at least one actuator in the directions of the interaction surface.

Images